An intelligent recommendation method and system for nursing plans based on knowledge graph
By calculating the similarity of clinical symptoms of patients in the knowledge graph and screening reference patients, combining nursing information and costs, the inherent problems of nursing plans in the existing technology are solved, and the intelligent recommendation of personalized nursing plans is realized, satisfying the uniqueness of patients. Demand and take into account economic factors.
Patent Information
- Application Number
- CN202411124803.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-16
AI Technical Summary
The existing intelligent recommendation methods for nursing programs based on knowledge graphs are difficult to meet the personalized needs of patients, resulting in the recommended nursing programs being too inherent and unable to fully meet the patients' unique conditions and needs.
By obtaining the differences in clinical information of patients in the knowledge graph, calculating the representation and similarity of clinical symptoms between patients, screening out the most similar reference patients, and combining the nursing information of these patients to establish a variety of recommendation plans, considering the nursing service. Applicability and cost, intelligent recommendation.
It realizes the recommendation of personalized nursing plans for recommended patients, provides the most applicable nursing services and corresponding levels, and takes into account the economic factors of the patients, expands the selection range and reference range, and improves the flexibility of recommendation.
Smart Images

Figure CN118645261B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for intelligently recommending nursing plans based on a knowledge graph. Background Art
[0002] The knowledge graph is a graphic that shows the development process and structural relationship of knowledge. It is applied to the intelligent transformation of industries related to nursing plan recommendation. By combining artificial intelligence and big data analysis, it can quickly recommend the best nursing plans and treatment strategies based on the patient's medical data and medical guidelines, thereby improving the accuracy and efficiency of clinical decision-making. It can also support nursing quality management, resource optimization and patient safety, becoming an indispensable and important part of modern medical services.
[0003] When making intelligent recommendations for nursing plans through big data analysis and knowledge graph technology, the various conditions of existing patients may often be compared with the patient information stored in the knowledge graph to obtain the similarity between the current patient and the stored information of each patient, and the nursing plan used by the patient with the highest similarity is applied to the current patient to achieve the recommendation of nursing plans. However, this may result in the nursing plans always being a few fixed plans and failing to meet the personalized needs of patients. Summary of the invention
[0004] The present invention provides a method and system for intelligently recommending nursing plans based on knowledge graphs to solve existing problems.
[0005] The present invention provides a method and system for intelligently recommending nursing plans based on knowledge graphs, which adopts the following technical solutions:
[0006] An embodiment of the present invention provides a method for intelligently recommending nursing plans based on a knowledge graph, the method comprising the following steps:
[0007] Use big data technology to collect nursing information related data and establish a knowledge graph. The nursing information related data should at least include: clinical information and nursing information of several patients. The clinical information includes several clinical symptoms and the corresponding disease severity. The patients include patients to be recommended. Two patients with a relationship in the knowledge graph are related patients.
[0008] According to the difference in clinical information of any two related patients, the representativeness of any clinical symptom of the same clinical symptoms of the two related patients to the patient is obtained; the representativeness of the same clinical symptoms of the two related patients to the two related patients is synthesized to obtain the preliminary similarity of the two related patients; the preliminary similarity between any two related patients and all other patients is screened to obtain similar patients of the two related patients; the preliminary similarity between all similar patients of any two related patients is synthesized to obtain the final similarity of the two related patients, and the reference patient of any patient is screened;
[0009] Summarize the nursing information of the reference patients of the patients to be recommended and establish several recommendation plans for the patients to be recommended;
[0010] Combining several recommended plans for the patients to be recommended, intelligent recommendation of nursing plans is completed.
[0011] Furthermore, the specific method of obtaining the representativeness of any clinical symptom of the same clinical symptoms of the two related patients to the patient according to the difference in clinical information of any two related patients includes:
[0012] Record any two related patients as patient A and patient B;
[0013]
[0014] in, Indicates the clinical symptoms that are common to patients A and B. The representativeness of each clinical symptom to patient A; represents the number of related patients of patient A; It means that patient A has the same clinical symptoms as patient A and patient B. The severity of clinical symptoms, Indicates the first The related patients have the same clinical symptoms as patients A and B. The severity of the clinical symptoms; represents the absolute value function.
[0015] Furthermore, the method of synthesizing the representativeness of the same clinical symptoms of two related patients to the two related patients to obtain the preliminary similarity of the two related patients includes the following specific methods:
[0016] Obtain the number of identical clinical symptoms in the clinical information of any two related patients;
[0017] The method for obtaining the preliminary similarity of any two related patients is:
[0018]
[0019] in, It indicates the preliminary similarity between patients A and B; It indicates the number of clinical symptoms that are common between patients A and B; , Indicates the clinical symptoms that are common to patients A and B. The representativeness of each clinical symptom to patients A and B; , Indicates that patient A and patient B have the same clinical symptoms. The severity of the clinical symptoms; represents the absolute value function, represents a normalization function, and T represents a preset first parameter.
[0020] Furthermore, the preliminary similarity between any two related patients and all other patients is screened to obtain similar patients of the two related patients, including the specific method of:
[0021] The preliminary similarity between any two related patients and all other patients is obtained, and patients whose preliminary similarity with the two related patients is greater than a preset second parameter are recorded as similar patients of the two related patients.
[0022] Furthermore, the method of synthesizing the preliminary similarities between all similar patients of any two related patients to obtain the final similarity of the two related patients and screening to obtain the reference patient of any patient includes the following specific methods:
[0023] The final similarity of any two related patients is obtained as follows:
[0024]
[0025] in, Indicates the final similarity between patient A and patient B; It indicates the preliminary similarity between patients A and B; represents the number of similar patients to patient A, represents the number of similar patients to patient B, Indicates the patient A The first similar patient and patient B Preliminary similarity of similar patients; represents the normalization function;
[0026] Among all the related patients of any patient, the related patients whose final similarity with the patient is greater than a preset second parameter are recorded as reference patients of the patient.
[0027] Furthermore, the nursing information of the reference patients of the patients to be recommended is integrated to establish several recommendation plans for the patients to be recommended, including the specific method of:
[0028] Obtaining nursing information of all reference patients of the patient to be recommended, wherein the nursing information includes nursing services, nursing service levels, and costs corresponding to the nursing service levels;
[0029] Obtain the applicability of any nursing service to the recommended patient; Obtain the applicability of any nursing service to the recommended patient;
[0030] Sort all nursing services in descending order of their applicability to the recommended patient. The nursing services and the corresponding applicable levels are recorded as recommended plan 1;
[0031] Obtain the priority of each nursing service being replaced and the priority of each nursing service level being replaced;
[0032] Sort each nursing service in order of priority from small to large, and replace the nursing service with the highest priority. The end of the nursing service and recommended plan The nursing service was replaced and the recommended solution 2 was obtained;
[0033] Sort each nursing service level in order of priority from small to large, and replace the one with the highest priority. Replace the applicable level of nursing service with the corresponding level of nursing service in recommended solution 1 to obtain recommended solution 3;
[0034] Said , , The fourth parameter, the fifth parameter and the sixth parameter are preset.
[0035] Furthermore, the specific method for obtaining the applicability of any nursing service to the recommended patient is:
[0036] For any nursing service, obtain the sum of the final similarities between all patients who received the nursing service among the reference patients of the patient to be recommended and patient A, and multiply it by the number of patients who received the nursing service among the reference patients of the patient to be recommended to obtain the applicability of the nursing service to the patient to be recommended.
[0037] Furthermore, the specific method for obtaining the applicable level of any nursing service to the recommended patient is:
[0038]
[0039] in, Indicates The appropriate level of nursing care for the recommended patient; Indicates the reference patients to be recommended who received Number of patients receiving each type of care; Indicates the reference patient to be recommended. The final similarity between the patient and the patient to be recommended, Indicates the reference patient to be recommended. The nursing service level of each patient receiving the hth type of nursing service; Indicates the floor symbol.
[0040] Furthermore, the specific method for obtaining the priority of each nursing service being replaced and the priority of each nursing service level being replaced is:
[0041] According to the cost of any nursing service for the applicable level of the recommended patient and the applicability of the nursing service to the recommended patient, the priority of each nursing service to be replaced is obtained; the cost of any nursing service for the applicable level of the recommended patient is directly proportional to the priority of the nursing service to be replaced, and the applicability of the nursing service to the recommended patient is inversely proportional to the priority of the nursing service to be replaced;
[0042] According to the cost of any nursing service level of any nursing service and the applicability of the nursing service to the recommended patients, the priority of each nursing service level being replaced is obtained; the cost of any nursing service level of any nursing service is directly proportional to the priority of the nursing service level being replaced, and the applicability of the nursing service to the recommended patients is inversely proportional to the priority of the nursing service level being replaced.
[0043] An embodiment of the present invention provides an intelligent recommendation system for nursing plans based on a knowledge graph, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of the method for intelligent recommendation of nursing plans based on a knowledge graph.
[0044] The beneficial effect of the technical solution of the present invention is as follows: the solution obtains patients who have a relationship with the patient to be recommended in the knowledge graph, determines the reference patient of the patient to be recommended according to the similarity of the degree of symptoms with the same clinical symptoms as the patient to be recommended, obtains the nursing service of the patient to be recommended according to the nursing information of the reference patient, and replaces the recommended nursing service with a more appropriate and less expensive nursing service taking into account the cost factor. The patient with the most similar symptoms to the patient to be recommended is screened out, and the nursing information of these patients is fully considered to obtain the most suitable nursing service and the corresponding level. At the same time, a variety of recommendation plans are provided, and the economic factors of the patient are considered when applicable, so that the patient has a wider range of choices, a larger reference range, and more flexibility. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0046] Figure 1 This is a flowchart of the steps of a method for intelligently recommending nursing plans based on a knowledge graph of the present invention. DETAILED DESCRIPTION
[0047] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of an intelligent recommendation method and system for nursing plans based on knowledge graphs proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0049] The specific scheme of the intelligent recommendation method and system for nursing plans based on knowledge graph provided by the present invention is described in detail below with reference to the accompanying drawings.
[0050] See also Figure 1 , which shows a flowchart of a method for intelligently recommending nursing plans based on a knowledge graph provided by an embodiment of the present invention, the method comprising the following steps:
[0051] Step S001: Use big data technology to collect nursing information related data and establish a knowledge graph.
[0052] Specifically, in order to implement the intelligent recommendation method for nursing plans based on knowledge graph proposed in this embodiment, it is first necessary to collect nursing information related data and establish a knowledge graph. The specific process is as follows:
[0053] Big data technology is used to collect nursing information-related data from several data sources and upload them to the database in a unified manner, and a knowledge graph is established for nursing-related data.
[0054] The nursing information related data should at least include: clinical information and nursing information of several patients; the clinical information includes several clinical symptoms and the corresponding disease severity, and the nursing information includes the nursing services selected by the patients, the corresponding nursing service levels and the corresponding fees.
[0055] It should be noted that the patients in the knowledge graph include patients to be recommended.
[0056] At this point, the knowledge graph of nursing information related data is obtained through the above method.
[0057] Step S002: According to the difference in clinical information of any two related patients, obtain the representativeness of any clinical symptom among the same clinical symptoms of the two related patients to the patient; obtain the preliminary similarity of the two related patients by combining the representativeness of the same clinical symptoms of the two related patients; screen the preliminary similarities between any two related patients and all other patients to obtain similar patients of the two related patients; combine the preliminary similarities between all similar patients of any two related patients to obtain the final similarity of the two related patients, and screen to obtain the reference patient of any patient.
[0058] It should be noted that in order to obtain the recommended nursing plan for the patient to be recommended, patients with similar symptoms to the patient to be recommended should be screened out from the knowledge graph, and the nursing plans selected by these patients should be used to recommend the patient to be recommended, so as to achieve a more suitable and effective treatment effect. Therefore, it is necessary to first obtain the reference patients of the patient to be recommended, and the basis for obtaining the reference patients is the correlation and similarity with the patient to be recommended in the knowledge graph.
[0059] Therefore, a preliminary screening is performed based on whether there is a relationship with the patient to be recommended in the knowledge graph, and then the similarity between the related patients of the patient to be recommended and the patient to be recommended is analyzed.
[0060] Specifically, in step 2.1, based on the difference in clinical information of any two related patients, the representativeness of any clinical symptom of the same clinical symptoms of the two related patients to the patient is obtained.
[0061] It should be noted that related patients are two patients who have a relationship in the knowledge graph.
[0062] It should be noted that when comparing the similarity between the current patient and the patients in the database, it is not reasonable to treat all clinical symptoms as parallel features and give them the same weight. Therefore, it is necessary to confirm the weight that each different clinical symptom should have when calculating the similarity based on the frequency of occurrence of each clinical symptom and the severity of the clinical symptoms. When a certain clinical symptom appears in more patients, it proves that its representativeness for the relevant patients is weaker, and the clinical symptom should have a lower weight when calculating the similarity; if a certain clinical symptom appears in any patient, the degree value is greatly different from that of other patients, which proves that the clinical symptom is more representative of the current patient, and the clinical symptom should have a higher weight when calculating the similarity.
[0063] As an example, any two related patients are recorded as patient A and patient B;
[0064]
[0065] in, Indicates the clinical symptoms that are common to patients A and B. The representativeness of each clinical symptom to patient A; represents the number of related patients of patient A; It means that patient A has the same clinical symptoms as patient A and patient B. The severity of clinical symptoms, Indicates the patient A The related patients have the same clinical symptoms as patients A and B. The severity of the clinical symptoms; represents the absolute value function.
[0066] It should be noted that, for any two related patients, the greater the difference between the severity of any same clinical symptom in any patient of the two related patients and the severity of the clinical symptom in other related patients of the patient, the more representative the clinical symptom is to the patient.
[0067] Step 2.2, integrating the representativeness of the same clinical symptoms of two related patients to the two related patients, and obtaining the preliminary similarity of the two related patients.
[0068] It should be noted that when calculating the preliminary similarity between two related patients, the number of identical clinical symptoms of the two related patients should be confirmed first. If there are no identical clinical symptoms, it means that the two related patients are not similar. On the basis of the identical clinical symptoms, the more identical clinical symptoms there are and the more similar the severity of the symptoms are, the higher the preliminary similarity between the two related patients. In addition, when calculating the preliminary similarity, it is necessary to take into account the representativeness of each clinical symptom to the patient being compared, so as to more accurately describe the preliminary similarity between the two related patients.
[0069] As an embodiment, the method for obtaining the preliminary similarity of any two related patients is:
[0070] Obtain the number of identical clinical symptoms in the clinical information of any two related patients;
[0071]
[0072] in, It indicates the preliminary similarity between patients A and B; It indicates the number of clinical symptoms that are common between patients A and B; , Indicates the clinical symptoms that are common to patients A and B. The representativeness of each clinical symptom to patients A and B; , Indicates that patient A and patient B have the same clinical symptoms. The severity of the clinical symptoms; represents the absolute value function, represents a normalization function, and T represents a preset first parameter.
[0073] It should be noted that It indicates the difference in the severity of the same clinical symptoms between patients A and B. The greater the difference, the lower the similarity between the two patients. It indicates the similarity of the degree of symptoms of patients A and B for the same clinical symptoms. The smaller the difference in the degree of symptoms, the greater the similarity of the degree of symptoms. All the common clinical symptoms of patients A and B are represented by taking the representativeness of each clinical symptom to patients A and B as weight, and the weighted sum of the similarities in the severity of each common clinical symptom of patients A and B is taken, and then multiplied by the number of common clinical symptoms between patients A and B to obtain the overall preliminary similarity between patients A and B.
[0074] It should be noted that the first parameter is preset based on experience It is 0.1 to avoid the situation where the denominator in the formula is 0. It can be adjusted according to actual conditions and is not specifically limited in this embodiment.
[0075] Step 2.3, screening the preliminary similarities between any two related patients and all other patients to obtain similar patients of the two related patients.
[0076] The preliminary similarity between any two related patients and all other patients is obtained, and patients whose preliminary similarity with the two related patients is greater than a preset second parameter are recorded as similar patients of the two related patients.
[0077] It should be noted that the second parameter is preset to 0.7 based on experience, and can be adjusted according to actual conditions, and is not specifically limited in this embodiment.
[0078] Step 2.4, synthesizing the preliminary similarities between all similar patients of any two related patients, obtaining the final similarity of the two related patients, and screening to obtain the reference patient of any patient.
[0079] It should be noted that, for two relatively similar related patients, the higher the initial similarity between other patients similar to the two patients, the higher the final similarity between the two patients should be. Therefore, the similarity between the two related patients can be corrected according to the initial similarity between the similar patients of the two related patients, and the final similarity between the two related patients can be obtained.
[0080] As an embodiment, the method for obtaining the final similarity of any two related patients is:
[0081]
[0082] in, Indicates the final similarity between patient A and patient B; It indicates the preliminary similarity between patients A and B; represents the number of similar patients to patient A, represents the number of similar patients to patient B, Indicates the first The first similar patient and patient B Preliminary similarity of similar patients; Represents the normalization function.
[0083] It should be noted that It means adding the preliminary similarities between all similar patients of two related patients to express the similarity between similar patients of two related patients. The more similar the similar patients of two related patients are, the more similar the two related patients are.
[0084] Among all related patients of any patient, the related patients whose final similarity with the patient is greater than a preset third parameter are recorded as reference patients of the patient.
[0085] It should be noted that the reference patient refers to a patient who has a high similarity with the patient to be recommended, and the nursing plan for the patient to be recommended can be determined based on the nursing plan of the reference patient.
[0086] So far, a reference patient for any patient is obtained through the above method.
[0087] Step S003: Summarize the nursing information of the reference patients of the patients to be recommended and establish several recommendation plans for the patients to be recommended.
[0088] It should be noted that when selecting a nursing plan for a patient to be recommended, the final similarity between the patient to be recommended and the reference patient should be considered, and various nursing services and their corresponding nursing service levels should be considered to obtain the most suitable recommendation plan for the patient to be recommended. At the same time, considering economic factors, the cost should also be taken into account, and relatively appropriate and low-cost nursing services and nursing service levels should be recommended to form a new recommendation plan.
[0089] Specifically, in step 3.1, the applicability of any nursing service to the patient to be recommended is obtained based on the final similarity between the reference patients of the patient to be recommended and the patient to be recommended, and the number of patients who received each nursing service among the reference patients.
[0090] Obtaining nursing information of all reference patients of the patient to be recommended, wherein the nursing information includes nursing services, nursing service levels, and costs corresponding to the nursing service levels;
[0091] For any nursing service, obtain the sum of the final similarities between all patients who received the nursing service among the reference patients of the patient to be recommended and patient A, and multiply it by the number of patients who received the nursing service among the reference patients of the patient to be recommended to obtain the applicability of the nursing service to the patient to be recommended.
[0092] Step 3.2, establish recommendation plan 1 based on the final similarity between the reference patients of the patient to be recommended and the patient to be recommended, the number of patients receiving each type of nursing service among the reference patients, and the level of each type of nursing service received.
[0093] It should be noted that the level of nursing services received by the reference patients of the recommended patient comprehensively reflects the level of nursing services that the recommended patient should receive. Therefore, for any nursing service, the average level of nursing services received by the reference patients is the recommended level of the recommended patient, and the more applicable it is to the recommended patient.
[0094] As an example, a specific method for obtaining the applicable level of any nursing service for a recommended patient is as follows:
[0095]
[0096] in, Indicates The appropriate level of nursing care for the recommended patient; Indicates the reference patients to be recommended who received Number of patients receiving each type of care; Indicates the reference patient to be recommended. The final similarity between the patient and the patient to be recommended, Indicates the reference patient to be recommended. The nursing service level of each patient receiving the hth type of nursing service; Indicates the floor symbol.
[0097] It should be noted that It means that the final similarity between the reference patient and the patient to be recommended is used as the weight, and the level of any nursing service selected by all reference patients is weighted averaged. The nursing service level selected by the reference patient with greater final similarity with the patient to be recommended has higher reference value, and the result is rounded down to finally obtain the applicable level of any nursing service for the patient to be recommended.
[0098] Sort all nursing services in descending order of their applicability to the recommended patient. The nursing service and the corresponding applicable level are recorded as recommended plan one.
[0099] Said It is the preset fourth parameter.
[0100] It should be noted that the fourth parameter is preset to 5 based on experience, and can be adjusted according to actual conditions, and is not specifically limited in this embodiment.
[0101] Step 3.3, based on the cost of the applicable level of any nursing service for the recommended patient and the applicability of the nursing service to the recommended patient, replace the recommended plan 1 and establish the recommended plan 2.
[0102] The priority of each nursing service to be replaced is obtained based on the cost of the applicable level of any nursing service for the recommended patients and the applicability of the nursing service to the recommended patients; the cost of the applicable level of any nursing service for the recommended patients is directly proportional to the priority of the nursing service to be replaced, and the applicability of the nursing service to the recommended patients is inversely proportional to the priority of the nursing service to be replaced.
[0103] Sort each nursing service in order of priority from small to large, and replace the nursing service with the highest priority. The end of the nursing service and recommended plan The nursing service is replaced and the recommended option 2 is obtained.
[0104] Said Preset the fifth parameter.
[0105] It should be noted that the fifth parameter is preset to 3 based on experience and can be adjusted according to actual conditions, and is not specifically limited in this embodiment.
[0106] Step 3.4, based on the cost of any nursing service level of any nursing service and the applicability of the nursing service to the recommended patient, replace the recommended plan one and establish the recommended plan three.
[0107] According to the cost of any nursing service level of any nursing service and the applicability of the nursing service to the recommended patients, the priority of each nursing service level being replaced is obtained; the cost of any nursing service level of any nursing service is directly proportional to the priority of the nursing service level being replaced, and the applicability of the nursing service to the recommended patients is inversely proportional to the priority of the nursing service level being replaced.
[0108] Sort each nursing service level in order of priority from small to large, and replace the one with the highest priority. Replace the applicable level of nursing service with the corresponding level of nursing service in recommended solution one to obtain recommended solution three.
[0109] Said Preset the sixth parameter.
[0110] It should be noted that the sixth parameter is preset to 3 based on experience, and can be adjusted according to actual conditions, and is not specifically limited in this embodiment.
[0111] So far, several recommendation plans for the patients to be recommended are obtained through the above method.
[0112] Step S004: Integrate several recommended plans for the patient to be recommended and complete the intelligent recommendation of the nursing plan.
[0113] Specifically, the applicability of the nursing services in several recommended plans, the applicable level of each nursing service and the cost are displayed to the recommended patient in the form of a visual diagram with reference to the patient's nursing information, and the nursing plan is finally determined.
[0114] It should be noted that entering the final care plan for the recommended patient into the knowledge graph can help the system to improve and upgrade itself, so that subsequent patients can receive more comprehensive and applicable care plans.
[0115] Through the above steps, intelligent recommendation of nursing plans based on knowledge graph is completed.
[0116] An embodiment of the present invention provides a knowledge graph-based intelligent nursing plan recommendation system, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of the knowledge graph-based intelligent nursing plan recommendation method.
[0117] This embodiment obtains patients who have a relationship with the patient to be recommended in the knowledge graph, determines the reference patient to be recommended based on the similarity of the degree of symptoms with the same clinical symptoms as the patient to be recommended, obtains the nursing service of the patient to be recommended based on the nursing information of the reference patient, and replaces the recommended nursing service with a more appropriate and less expensive nursing service taking into account the cost factor. The patients with the most similar symptoms to the patient to be recommended are screened out, and their nursing information is fully considered to obtain the most suitable nursing service and corresponding level. At the same time, a variety of recommendation plans are provided, and the economic factors of the patient are considered when applicable, so that the patient has a wider range of choices, a larger reference range, and more flexibility.
[0118] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for intelligently recommending nursing plans based on knowledge graph, characterized in that: The method comprises the following steps: Use big data technology to collect nursing information related data and establish a knowledge graph. The nursing information related data should at least include: clinical information and nursing information of several patients. The clinical information includes several clinical symptoms and the corresponding disease severity. The patients include patients to be recommended. Two patients with a relationship in the knowledge graph are related patients. According to the difference in clinical information of any two related patients, the representativeness of any clinical symptom of the same clinical symptoms of the two related patients to the patient is obtained; the representativeness of the same clinical symptoms of the two related patients to the two related patients is synthesized to obtain the preliminary similarity of the two related patients; the preliminary similarity between any two related patients and all other patients is screened to obtain similar patients of the two related patients; the preliminary similarity between all similar patients of any two related patients is synthesized to obtain the final similarity of the two related patients, and the reference patient of any patient is screened; Summarize the nursing information of the reference patients of the patients to be recommended and establish several recommendation plans for the patients to be recommended; Comprehensively propose several recommended plans for the patients to be recommended, and complete the intelligent recommendation of nursing plans; The specific method of obtaining the representativeness of any clinical symptom of the same clinical symptoms of the two related patients to the patient according to the difference in clinical information of the two related patients includes: Record any two related patients as patient A and patient B; in, Indicates the clinical symptoms that are common to patients A and B. The representativeness of each clinical symptom to patient A; represents the number of related patients of patient A; It means that patient A has the same clinical symptoms as patient A and patient B. The severity of clinical symptoms, Indicates the patient A The related patients have the same clinical symptoms as patients A and B. The severity of the clinical symptoms; represents the absolute value function; The method of synthesizing the representativeness of the same clinical symptoms of two related patients to the two related patients and obtaining the preliminary similarity of the two related patients includes the following specific methods: Obtain the number of identical clinical symptoms in the clinical information of any two related patients; The method for obtaining the preliminary similarity of any two related patients is: in, It indicates the preliminary similarity between patients A and B; It indicates the number of clinical symptoms that are common between patients A and B; , Indicates the clinical symptoms that are common to patients A and B. The representativeness of each clinical symptom to patients A and B; , Indicates that patient A and patient B have the same clinical symptoms. The severity of the clinical symptoms; represents the absolute value function, represents a normalization function, and T represents a preset first parameter; The specific method of synthesizing the nursing information of the reference patients of the patients to be recommended and establishing several recommendation plans for the patients to be recommended includes: Obtaining nursing information of all reference patients of the patient to be recommended, wherein the nursing information includes nursing services, nursing service levels, and costs corresponding to the nursing service levels; Obtain the applicability of any nursing service to the recommended patient; Obtain the applicability of any nursing service to the recommended patient; Sort all nursing services in descending order of their applicability to the recommended patient. The nursing services and the corresponding applicable levels are recorded as recommended plan 1; Obtain the priority of each nursing service being replaced and the priority of each nursing service level being replaced; Sort each nursing service in order of priority from small to large, and replace the nursing service with the highest priority. The end of the nursing service and recommended plan The nursing service was replaced and the recommended solution 2 was obtained; Sort each nursing service level in order of priority from small to large, and replace the one with the highest priority. Replace the applicable level of nursing service with the corresponding level of nursing service in recommended solution 1 to obtain recommended solution 3; Said , , The fourth parameter, the fifth parameter and the sixth parameter are preset.
2. According to claim 1, a method for intelligently recommending nursing plans based on knowledge graphs is characterized in that: The screening of the preliminary similarities between any two related patients and all other patients to obtain similar patients of the two related patients includes the following specific methods: The preliminary similarity between any two related patients and all other patients is obtained, and patients whose preliminary similarity with the two related patients is greater than a preset second parameter are recorded as similar patients of the two related patients.
3. According to claim 1, a method for intelligently recommending nursing plans based on knowledge graph, characterized in that: The method of synthesizing the preliminary similarities between all similar patients of any two related patients to obtain the final similarity of the two related patients and screening to obtain the reference patients of any patients includes the following specific methods: The final similarity of any two related patients is obtained as follows: in, Indicates the final similarity between patient A and patient B; It indicates the preliminary similarity between patients A and B; represents the number of similar patients to patient A, represents the number of similar patients to patient B, Indicates the patient A The first similar patient and patient B Preliminary similarity of similar patients; represents the normalization function; Among all the related patients of any patient, the related patients whose final similarity with the patient is greater than a preset second parameter are recorded as reference patients of the patient.
4. According to claim 1, a method for intelligently recommending nursing plans based on knowledge graph, characterized in that: The specific method for obtaining the applicability of any nursing service to the recommended patient is as follows: For any nursing service, obtain the sum of the final similarities between all patients who received the nursing service among the reference patients of the patient to be recommended and patient A, and multiply it by the number of patients who received the nursing service among the reference patients of the patient to be recommended to obtain the applicability of the nursing service to the patient to be recommended.
5. According to claim 1, a method for intelligently recommending nursing plans based on knowledge graph, characterized in that: The specific method for obtaining the applicable level of any nursing service for the recommended patient is as follows: in, Indicates The appropriate level of nursing care for the recommended patient; Indicates the reference patients to be recommended who received Number of patients receiving each type of care; Indicates the reference patient to be recommended. The final similarity between the patient and the patient to be recommended, Indicates the reference patient to be recommended. The nursing service level of each patient receiving the hth type of nursing service; Indicates the floor symbol.
6. According to claim 1, a method for intelligently recommending nursing plans based on knowledge graph, characterized in that: The specific method for obtaining the priority of each nursing service being replaced and the priority of each nursing service level being replaced is: According to the cost of any nursing service for the applicable level of the recommended patient and the applicability of the nursing service to the recommended patient, the priority of each nursing service to be replaced is obtained; the cost of any nursing service for the applicable level of the recommended patient is directly proportional to the priority of the nursing service to be replaced, and the applicability of the nursing service to the recommended patient is inversely proportional to the priority of the nursing service to be replaced; According to the cost of any nursing service level of any nursing service and the applicability of the nursing service to the recommended patients, the priority of each nursing service level being replaced is obtained; the cost of any nursing service level of any nursing service is directly proportional to the priority of the nursing service level being replaced, and the applicability of the nursing service to the recommended patients is inversely proportional to the priority of the nursing service level being replaced.
7. A knowledge graph-based intelligent nursing plan recommendation system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.
